python - Theano is claiming computation graph does not need the most important variable -
prediction = t.argmax(t.nnet.softmax(w2.dot(t.tanh(w1.dot(x))))); cost = ((prediction - y)**2).sum(); dw1 = t.grad(cost, w1); dw1 = dw1.eval({ w1 : w1, x: data, y : labels }); i error:
unusedinputerror: theano.function asked create function computing outputs given inputs, provided input variable @ index 1 not part of computational graph needed compute outputs: y. this ridiculous. y part of computation graph.
why theano doing this?
Comments
Post a Comment